Dissecting the shared genetic architecture of adipose tissue aging and aging-related lipid-metabolic dysfunction: a multivariate genomic analysis

Abstract

Adipose tissue aging is increasingly recognized as a systemic process involving altered fat distribution, ectopic lipid deposition, and metabolic dysfunction, yet its shared genetic architecture remains incompletely characterized. In this study, we constructed a latent multivariate genetic phenotype of adipose tissue aging by integrating genome-wide association summary statistics for seven adipose- and metabolism-related traits, including visceral adipose tissue volume, abdominal subcutaneous adipose tissue volume, thigh subcutaneous adipose tissue volume, liver proton density fat fraction, fasting insulin, adiponectin, and leptin. Using Genomic Structural Equation Modelling, we derived a latent phenotype termed mvAdiposeAge and further characterized its genetic and biological relevance through genome-wide association analysis, FUMA-based functional annotation, MAGMA gene-level analysis, transcriptome-wide association analysis, expression fine-mapping, pathway and tissue enrichment analyses, and Mendelian randomization. The Genomic SEM model supported a shared genetic dimension linking adipose distribution, ectopic lipid accumulation, and adipose-related metabolic traits. Genome-wide analysis identified a prominent signal at chromosome 16q12.2, with rs1421085 in the FTO region emerging as the lead variant. MAGMA identified 49 FDR-significant genes, while transcriptome-wide analysis prioritized 11 genes, with FOCUS fine-mapping highlighting expression-mediated signals involving FTO, RPGRIP1L, and AKTIP. Functional enrichment implicated neurotrophin signalling, synaptic organization, cAMP signalling, immune and phagocytic processes, endocrine regulation, and metabolic pathways, with tissue enrichment suggesting prominent brain-region-specific signals. Mendelian randomization further supported positive associations of adiposity-related traits with mvAdiposeAge and an inverse association with cognitive performance. These findings highlight mvAdiposeAge as a latent genetic factor capturing shared genetic variation across adipose-related phenotypes relevant to aging and provide a framework for further investigating their systemic biological associations.